Thanks for the feedback! I would guess you are high in straightforwardness 
.
I feel like the test is too stark.
Could you elaborate? Would you prefer it to have more decoration and color?
The icons are sort of stupid imho.
They look cheap.
It looks better at the smaller size, so I’d suggest limiting to that size window myself.
We have mixed feedback on this; I appreciate the datapoint. Since we’re dealing with such a broad population, sometimes I feel our design panders to the mean instead of choosing a strong opinion.
The end should have an ending screen
Yeah, good idea!
basically it all depends on how accurate your assessment is. The test itself is pretty vanilla
Good points, accuracy is two parts: 1) the test takers mentality and 2) is hidden in how we key the actual results. The normalization happens behind the scenes where you’re really answering the same question multiple times in different ways to determine a moving average.
Over time we want to gather more data from a longer time series. This ties into the “there’s a lot of questions” comment, which is to say it would be better to spread them out over time. As I mentioned, the actual measured “accuracy” across the entire dataset could be thought of as 87%. Meaning it’s not really going to change more than 13% if you take it again barring retester liability.
This is a useful but technically inaccurate statement. It is more correct to say that it measures what it intends to measure the large majority of the time, which can be a proxy for accuracy. It is highly reliable in many directions and valid due to the factor analysis that led to the initial attributes measured.